PAI-1-IL-17 statistical interaction and periventricular white matter hyperintensity severity in sporadic cerebral small vessel disease.

Abstract

The factors influencing the severity of periventricular white matter hyperintensity (WMH) in cerebral small vessel disease (CSVD) are not fully understood. This study investigated the multiplicative interaction between circulating plasminogen activator inhibitor-1 (PAI-1) and interleukin-17 (IL-17) concerning periventricular WMH severity. In this single-center, cross-sectional study, we enrolled 151 patients with sporadic CSVD, selected from 526 consecutively screened inpatients. Serum levels of PAI-1, IL-17, and matrix metalloproteinase-9 (MMP-9) were quantified using enzyme-linked immunosorbent assays. Regional WMH was graded on magnetic resonance imaging using the Fazekas scale. Multivariable logistic regression was employed to assess the PAI-1-IL-17 interaction for moderate-to-severe periventricular WMH (Fazekas scores 2-3 vs. 0-1), with additional adjustments for MMP-9 and sensitivity analyses. The primary interaction odds ratio (OR) was 1.72 per one-standard-deviation increase in the standardized product term (95% confidence interval [CI]: 1.07-2.77; p = 0.026). Adjusting for MMP-9 did not substantially change this estimate. The interaction was not statistically significant in the ordinal periventricular analysis (OR: 1.40; 95% CI: 0.99-1.99; p = 0.061) or the deep-WMH model (OR: 0.83; 95% CI: 0.51-1.37; p = 0.468). While alternative binary models showed positive interaction estimates, resampling analyses indicated limited stability, and additive interaction lacked consistent support. These exploratory findings suggest a multiplicative statistical interaction between PAI-1 and IL-17 for dichotomized periventricular WMH severity. However, they do not establish causality, biological synergy, or clinical biomarker utility. Prospective multicenter validation is essential.